This is a test version of Biostars. For the public version, visit https://www.biostars.org.
Looking For Network Visualization Tools

Hi to all. I would like to plot a network of interacting genes. I performed Mutual Information between genes starting from gene expression values, so my adjacency matrix is composed of mutual information values. Now I would like to have a visualization of the interactions. I tried to use Cytoscape or R igraph or RCytoscape but all this methods transform the adjacency matrix in a boolean matrix. Is there a way or a tool that does not transform the adj matrix in a boolean one? This because according to the mutual information value I would like to have different lengths in edges. For example if A and B interact with a mutual information value of 0.20, the edge connecting them will be longer than the edge connecting A and C that show a mutual information of 0.1.

Any suggestion about this?

Thank you in advance

network plot
• 9,274 views • link
updated by Ram Ram • written by elb elb

4 answers

You might be interested in CLANS software. It performs clustering (2D or 3D) based on distance matrix (exactly what you have) and has a support for microarray experiments.

CLANS screenshot

• 120 views • link
written by Pawel Szczesny Pawel Szczesny

Hi Pawel!! It is a fantastic tool! thank you sooo much!!

• 0 views • link
written by elb elb

Sounds like you have a distance matrix rather than an adjacency matrix, I've not used cytoscape but it looks like it's easy to make a graph with weighted edges by modifying your input format slightly (link).

• 2 views • link
written by Ben Ben

Hi Ben!! yes, I suppose it is more like a distance matrix than an .adj matrix!

• 0 views • link
written by elb elb

I have used similar data without a problem with igraph. Try:

library(igraph)
g <- graph.adjacency(as.matrix(yourData), mode="undirected", weighted=TRUE)
• 0 views • link
written by fanofactor • fanofactor

Hi fanofactor!!! Thank you for your help! Unfortunately while I solved the problem of MI values setting weighted = T, I encountered another problem because that function (graph.adjacency) needs a squared matrix, and my matrix is not a squared one..

• 0 views • link
written by elb elb

Try Gephi - an interactive visualization and exploration platform for all kinds of networks and complex systems, dynamic and hierarchical graphs. https://gephi.org/

• 1 views • link
written by zx8754 zx8754

Hi zx8754! What a fantastic tool!!!

• 0 views • link
written by elb elb

Log in to answer this question.

More posts like this